Postgraduate Diploma in Segmentation/Detection with AI and Validation
Online Postgraduate Diploma of 800 hours in Segmentation/Detection with AI and Validation. 8 modules covering Regulation (EU) 2024/1689 (Artificial Intelligence), Regulation (EU) 2017/745 (MDR) and MONAI. Assigned tutor and open enrolment.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Describe the formats and characteristics of digital medical imaging.
Organise annotation and the reference standard.
Train segmentation architectures.
Develop detection and classification models.
Validate models with appropriate technical metrics.
Design clinical validation studies.
Prepare deployment in compliance with the MDR and the AI Regulation.
Who it's for
- Medical computer vision engineer who wishes to specialise in Segmentation/Detection with AI and Validation.
- Imaging data scientist who wishes to specialise in Segmentation/Detection with AI and Validation.
- Clinical AI validation specialist who wishes to specialise in Segmentation/Detection with AI and Validation.
- AI regulatory affairs engineer who wishes to specialise in Segmentation/Detection with AI and Validation.
- Deep learning researcher who wishes to specialise in Segmentation/Detection with AI and Validation.
- Healthcare MLOps engineer who wishes to specialise in Segmentation/Detection with AI and Validation.
- Engineering or science graduates looking to enter hospitals.
What you'll work with
- Regulation (EU) 2024/1689 (Artificial Intelligence)
- Regulation (EU) 2017/745 (MDR)
- IEC 62304 (medical device software)
- ISO/IEC 23894 (AI risk management)
- DICOM
- CLAIM guideline (AI in medical imaging)
- ISO 14971 (risk management)
8 modules · 800 h
1.1Modalities: radiography, CT, MRI, ultrasound and PET1.2DICOM standard and metadata1.3Hounsfield units and display windows1.4Reconstruction and voxel spacing1.5Anonymisation of DICOM studies1.6Resampling and intensity normalisation1.7Public datasets: LIDC-IDRI, BraTS and MSD1.8Tools: 3D Slicer and ITK-SNAP1.9NIfTI formats and conversion1.10Practical: pre-processing volumes with SimpleITK
2.1Annotation protocols by structure2.2Inter- and intra-observer variability2.3Fusion of annotations with STAPLE2.4Assisted annotation with MONAI Label2.5Weak and image-level labelling2.6Quality control of masks2.7Expert consensus and adjudication2.8Cost and scaling of annotation2.9Documentation of the reference standard2.10Workshop: liver lesion annotation protocol
3.1U-Net and 3D U-Net3.2Automatic configuration with nnU-Net3.3Attention and residual connections3.4Transformers: UNETR and Swin UNETR3.5Foundation models: SAM and MedSAM3.6Loss functions: Dice, focal and combined3.7Data augmentation in medical imaging3.8Patch-based training and sliding window3.9Connected-component postprocessing3.10Practical: training nnU-Net on a public dataset
4.1Detectors: Faster R-CNN, RetinaNet and YOLO4.2Detection of pulmonary nodules4.3Computer-aided detection (CADe) versus diagnosis (CADx)4.4Mammography and tomosynthesis4.5Activation maps and Grad-CAM4.6Multitask detection-segmentation learning4.7Detection in endoscopic video4.8Multiple-instance classification in pathology4.9Selection of operational thresholds4.10Laboratory: nodule detector with FROC analysis
5.1Dice, IoU and 95th percentile Hausdorff distance5.2Metrics Reloaded recommendations5.3Sensitivity, specificity, AUC and FROC5.4Bootstrap confidence intervals5.5Patient-level splitting and data leakage5.6Multicentre external validation5.7Calibration of probabilities5.8Robustness across different scanners and protocols5.9Subgroup error analysis5.10Practical: validation report with confidence intervals
6.1Standalone versus assisted performance6.2MRMC reader studies6.3Design of diagnostic accuracy studies6.4CLAIM checklist6.5DECIDE-AI guideline for early evaluation6.6Automation bias in radiologists6.7Impact on workflow and turnaround times6.8Prospective studies and clinical impact6.9Economic evaluation of deployment6.10Case study: designing an MRMC study
7.1Classification under the MDR7.2510(k) pathway for CADe and CADx at the FDA7.3Risk management under ISO 149717.4Software under IEC 623047.5Predetermined change control plans (PCCP)7.6PACS integration with DICOM SEG and SR7.7Deployment with MONAI Deploy7.8Drift monitoring in production7.9AI Regulation and high risk7.10Workshop: technical file for a segmentation algorithm
8.1Definition of the clinical problem8.2Selection and preparation of the dataset8.3Annotation protocol8.4Model training8.5Technical validation with segmentation metrics8.6Clinical validation plan8.7Regulatory and integration strategy8.8Analysis of limitations8.9Report and presentation8.10Final review with the tutor
How you'll learn
100% online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. No fixed timetable: students progress at their own pace within the intake, which makes it possible to combine study with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is MONAI, nnU-Net, PyTorch, 3D Slicer, ITK-SNAP, in free, academic or evaluation versions. Trials are studied using real dataset annotation procedures with a reference standard, technical validation with Dice and Hausdorff metrics, and reader-algorithm studies. Assessment is continuous per module, plus the final integrating project.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Describe the formats and characteristics of digital medical imaging.
Organise annotation and the reference standard.
Train segmentation architectures.
Develop detection and classification models.
Validate models with appropriate technical metrics.
Where this takes you
Requirements
- Prior technical training or professional experience in the sector. A university degree is not required.
- Internet connection and a computer capable of running the programme's software.
- Sufficient level of Spanish to follow technical regulatory texts.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme totals 800 hours spread over 6 to 8 months. That is between 10 and 15 hours per week on average, compatible with a full-time job.
You work with MONAI, nnU-Net, PyTorch, 3D Slicer, ITK-SNAP. Free, academic or evaluation versions are used, and the tutor indicates the free alternative where one exists.
SEIUM's own diploma, awarded on passing the continuous assessment and the integrating project. This is in-house training, not an official qualification from the Ministry.
Yes. Modules passed in a course or postgraduate diploma in the same area are recognised in the advanced programme. Check the recognition policy when you enrol.
Related programmes
Postgraduate Diploma in Life Cycle Testing and Electrical Safety
Online 800-hour Postgraduate Diploma in Life Cycle Testing and Electrical Safety. 8 modules covering IEC 60601-1 (safety of medical electrical equipment), IEC 62368-1 (safety of ICT…
Postgraduate Diploma in Motion Capture and Haptic Feedback
Online Postgraduate Diploma of 800 h in Motion Capture and Haptic Feedback. 8 modules covering ISO 9241-910 (touch and haptic interaction, framework), ISO 9241-920 (guidance on…
Postgraduate Diploma in Cabin Hazards: Ions, Noise and Vibration
Online 800-hour Postgraduate Diploma in Cabin Hazards: Ions, Noise and Vibration. 8 modules covering Directive 2013/59/Euratom (cosmic radiation for crews), Directive 2003/10/EC (noise) and CARI-7 (FAA).…
More in Health, Biomechanics, Wearables & Operational Safety
All Health, Biomechanics, Wearables & Operational Safety programmes → · How to apply · Fees & financing · Scholarships